Table 5 Correlation between deep learning derived substitution matrices and benchmarking matrices
From: 3D deep convolutional neural networks for amino acid environment similarity analysis
Matrix 1 | Matrix 2 | R-value |
|---|---|---|
3DCNN- S freq | BLOSUM62 | 0.702 |
3DCNN- S freq | PAM250 | 0.609 |
3DCNN- S freq | WAC | 0.613 |
3DCNN- S dot | BLOSUM62 | 0.82 |
3DCNN- S dot | PAM250 | 0.788 |
3DCNN- S dot | WAC | 0.6183 |
WAC | BLOSUM62 | 0.811 |
WAC | PAM250 | 0.681 |
PAM250 | BLOSUM62 | 0.872 |